betaFeedback

WorksMathematics

Gibbs measures on Cayley trees and homoclinic orbits

Abstract

We study the question of existence and multiplicity of homogeneous Markov Gibbs measures on Cayley trees with spin space Z\mathbb{Z}. We prove that, under a natural summability assumption on the interaction function, there are always at least two homogeneous Markov Gibbs measures satisfying two natural symmetry conditions. Moreover, any two homogeneous Markov Gibbs measures satisfying a suitable non-degeneracy assumption can be superimposed and produce infinitely many other homogeneous Markov Gibbs measures. When the interaction function QQ has bounded support in Z\mathbb{Z}, there is a one-to-one correspondence between homogeneous Markov Gibbs measures and suitable homoclinic orbits of a symplectomorphism of R2n\mathbb{R}^{2n}, where nn is the smallest natural number such that Q(i)=0Q(i)=0 for every i>ni>n. For n=1n=1, results from symbolic dynamics and horseshoes imply the existence of homogeneous Markov Gibbs measures whose single-site marginal is concentrated near an arbitrary finite subset of NZN\mathbb{Z}, where NN is a sufficiently large integer. In the case of interaction functions with infinite support, there is a one-to-one correspondence between homogeneous Markov Gibbs measures and suitable homoclinic orbits of a Lagrangian correspondence on the cotangent bundle of a Banach space.

Introduction

Homogeneous Markov Gibbs measures on Cayley trees. This paper addresses the questions of existence and multiplicity of homogeneous Markov Gibbs measures on Cayley trees, in the case of an infinite spin space. We recall that the Cayley tree of order d≥2d \ge2 is the unique unrooted tree with countably many vertices, each of which is the origin of d+1d+1 edges. We denote by VV and EE the sets of its vertices and edges. We choose as spin space the ring of integers Z\mathbb{Z}, thus considering Ω=ZV\Omega= \mathbb{Z}^{V} as configuration space, endowed with its usual σ\sigma-algebra A\mathcal{A}.

A probability measure on (Ω,A)(\Omega,\mathcal{A}) is said to be homogeneous if it is invariant under all tree automorphisms, and a homogeneous probability measure μ\mu is said to be Markov if it is determined by its single-site marginal π:Z→[0,1]\pi:\mathbb{Z}\to[0,1] and transition matrix P:Z×Z→[0,1]P:\mathbb{Z}\times\mathbb{Z}\to[0,1] by propagation along the tree, where

π(i):=μ(σx=i),P(i,j):=μ(σy=j∣σx=i),∀i,j∈Z.\pi(i):=\mu(\sigma_x=i),\qquad P(i,j):=\mu(\sigma_y=j\mid\sigma_x=i),\qquad\forall i,j\in\mathbb{Z}.

Here, σx\sigma_x denotes the value of the configuration σ∈Ω\sigma\in\Omega at x∈Vx\in V, {x,y}\{x,y\} is an edge, and the functions π\pi and PP do not depend on xx and {x,y}\{x,y\} because of homogeneity.

In order to define Gibbs measures, we need an interaction function Q:Z→RQ:\mathbb{Z}\to\mathbb{R}, which we assume to be nonnegative, even, not identically zero, and d+12\frac{d+1}{2}-summable. Throughout the introduction, we shall also assume QQ to be irreducible, meaning that the abelian subgroup of Z\mathbb{Z} generated by its support is the whole Z\mathbb{Z}. Typical and well-studied examples are the interaction functions

Q(i)=e−∣i∣T,T>0,Q(i)=(11+∣i∣)1T,0<T<d+12,∀i∈Z,(1)Q(i)=e^{-\frac{|i|}{T}},\quad T>0,\qquad Q(i)=\left(\frac{1}{1+|i|}\right)^{\frac{1}{T}},\quad0<T<\frac{d+1}{2},\qquad\forall i\in\mathbb{Z}, \tag*{(1)}

where in physical applications the positive parameter TT is interpreted as temperature. The Gibbs property for a homogeneous Markov probability measure μ\mu on (Ω,A)(\Omega,\mathcal{A}) can be expressed in the following way: given a vertex x∈Vx \in V with neighbours y1,…,yd+1y_1,\ldots,y_{d+1}, we have

μ(σx=i∣σy1=j1,…,σyd+1=jd+1)=∏h=1d+1Q(i−jh)∑k∈Z∏h=1d+1Q(k−jh).\mu(\sigma_x=i\mid\sigma_{y_1}=j_1,\ldots,\sigma_{y_{d+1}}=j_{d+1})=\frac{\prod_{h=1}^{d+1}Q(i-j_h)}{\sum_{k\in\mathbb{Z}}\prod_{h=1}^{d+1}Q(k-j_h)}.

for every i∈Zi \in\mathbb{Z} and (j1,…,jd+1)∈Zd+1(j_1,\ldots,j_{d+1}) \in\mathbb{Z}^{d+1} for which the conditioning event has positive probability. See § 2 for more precise definitions and further discussion. See also the classical book of Georgii [5] for the general theory of Gibbs measures on graphs and the book of Rozikov [14], focusing on models on Cayley trees.

The convolution equation. We denote by G(d,Q)\mathcal{G}(d,Q) the set of homogeneous Markov probability measures on the configuration space Ω\Omega associated to the Cayley tree of order dd which are Gibbs with respect to the interaction function QQ. Here it is worth noticing that the Gibbs property depends only on the projective class of QQ, meaning that if cc is any positive number then G(d,cQ)=G(d,Q)\mathcal{G}(d,cQ)=\mathcal{G}(d,Q). Note also that Z\mathbb{Z} acts by translation on the spin space and hence on G(d,Q)\mathcal{G}(d,Q). Our interest is in studying the number of elements of the quotient G(d,Q)/Z\mathcal{G}(d,Q)/\mathbb{Z} by this Z\mathbb{Z}-action and their qualitative properties.

Due to the irreducibility of QQ, there is a one-to-one correspondence between the Gibbs measures in G(d,Q)\mathcal{G}(d,Q) and the positive solutions u∈ℓp(Z)u \in\ell^p(\mathbb{Z}), p:=d+1dp:=\frac{d+1}{d}, of the equation

u=(Q∗u)d(2)u=(Q*u)^d \tag*{(2)}

where the symbol ∗* denotes the convolution product of two functions on Z\mathbb{Z} and ℓp(Z)\ell^p(\mathbb{Z}) the space of pp-summable bi-infinite real sequences. The single-site marginal and transition matrix of the Gibbs measure associated to the positive solution u∈ℓp(Z)u \in\ell^p(\mathbb{Z}) of (2) are given by the identities

π(i)=u(i)p∥u∥pp,P(i,j)=Q(i−j)u(j)(Q∗u)(i).\pi(i)=\frac{u(i)^p}{\|u\|_p^p},\qquad P(i,j)=\frac{Q(i-j)u(j)}{(Q*u)(i)}.

where ∥u∥p\|u\|_p denotes the pp-norm of uu. This fundamental fact, which was proven in a much more general form by Zachary in [17], is reproven for the reader’s convenience in § 2.

We call the positive number

w:=∥Q∥d+12dd+1∥u∥p(3)w:=\|Q\|_{\frac{d+1}{2}}^{\frac{d}{d+1}}\|u\|_p \tag*{(3)}

weight of the Gibbs measure μ\mu associated to the solution uu to (2). Here, the exponent in the norm of QQ is chosen in such a way that ww does not change if we replace QQ by cQcQ for some c>0c>0. (2) and Young’s inequality imply that w≥1w\geq1. Weights play an important role in Theorem B below.

As far as we know, up to now equation (2) has been studied mainly in a perturbative regime, see [6] and [1]. If Q=1{0}Q=\mathbb{1}_{\{0\}} is the characteristic function of the singleton {0}\{0\} then (2) reduces to the equation u=udu=u^d, whose nonnegative solutions in ℓp(Z)\ell^p(\mathbb{Z}) are all the characteristic functions of finite subsets of Z\mathbb{Z}. In the zero-temperature limit T→0T\to0, interaction functions QQ such as those defined in (1) converge to 1{0}\mathbb{1}_{\{0\}} and one expects to find positive solutions of (2) in ℓp(Z)\ell^p(\mathbb{Z}) which are close to characteristic functions of finite subsets of Z\mathbb{Z}. Indeed, in [1 Theorem 3.1] it is proven that for every natural number NN there exists a positive threshold η=η(d,N)\eta=\eta(d,N) such that if QQ is normalized by Q(0)=1Q(0)=1 and ∥Q−1{0}∥d+12<η\lVert Q-\mathbf{1}_{\{0\}}\rVert_{\frac{d+1}{2}}<\eta then G(d,Q)\mathcal{G}(d,Q) contains Gibbs measures whose single-site marginals are concentrated near arbitrary subsets of Z\mathbb{Z} of cardinality at most NN.

In this paper, we wish to derive existence and multiplicity results outside of the perturbative regime. In all our results, the starting point is equation (2). Therefore, this paper can also be considered as an investigation of several techniques for studying this fixed point equation. From this point of view, the assumption that dd is an integer is not necessary and all our results concerning solutions of (2) would hold with minor changes for any real number d>1d>1.

Existence of two symmetric Gibbs measures. Consider the involutions ι,ȷ:Z→Z\iota,\jmath:\mathbb{Z}\to\mathbb{Z} defined by ι(i)=−i\iota(i)=-i and ȷ(i)=1−i\jmath(i)=1-i, which lift to involutions on Ω=ZZ\Omega=\mathbb{Z}^{\mathbb{Z}}, and by the push-forward operation on measures to involutions ι∗\iota_* and ȷ∗\jmath_* on the space of probability measures on (Ω,A)(\Omega,\mathcal{A}). The fact that QQ is even implies that the set of Gibbs measures G(d,Q)\mathcal{G}(d,Q) is invariant under both ι∗\iota_* and ȷ∗\jmath_*. It then makes sense to look for Gibbs measures in G(d,Q)\mathcal{G}(d,Q) which are fixed by either ι∗\iota_* or ȷ∗\jmath_*. Equivalently, we are looking for Gibbs measures μ\mu in G(d,Q)\mathcal{G}(d,Q) whose single-site marginal π\pi and transition matrix PP satisfy either

π(−i)=π(i),P(−i,−j)=P(i,j)∀(i,j)∈Z×Z.\pi(-i)=\pi(i), \qquad P(-i,-j)=P(i,j) \qquad\forall(i,j)\in\mathbb{Z}\times\mathbb{Z}.

or

π(1−i)=π(i),P(1−i,1−j)=P(i,j)∀(i,j)∈Z×Z.\pi(1-i)=\pi(i), \qquad P(1-i,1-j)=P(i,j) \qquad\forall(i,j)\in\mathbb{Z}\times\mathbb{Z}.

The first new result of this paper is the following theorem, stating that Gibbs measures which are symmetric in the above two senses always exist.

Theorem A. Assume that the interaction function QQ is irreducible and d+12\frac{d+1}{2}-summable. Then there exist Gibbs measures μ1\mu_1 and μ2\mu_2 in G(d,Q)\mathcal{G}(d,Q) such that ι∗(μ1)=μ1\iota_*(\mu_1)=\mu_1 and ȷ∗(μ2)=μ2\jmath_*(\mu_2)=\mu_2. In particular, μ1\mu_1 and μ2\mu_2 define two distinct elements of G(d,Q)/Z\mathcal{G}(d,Q)/\mathbb{Z}.

The proof of this theorem is variational: we find pp-summable positive solutions uu of (2) by maximizing the functional

Q(u):=⟨Q∗u,u⟩=∑(i,j)∈Z2Q(j−i)u(i)u(j)\mathcal{Q}(u):=\langle Q*u,u\rangle=\sum_{(i,j)\in\mathbb{Z}^2}Q(j-i)u(i)u(j)

on the unit sphere in ℓp(Z)\ell^p(\mathbb{Z}). The symmetry conditions are achieved by constraining this maximization to bi-infinite sequences uu which satisfy either u(−i)=u(i)u(-i)=u(i) or u(1−i)=u(i)u(1-i)=u(i), for every i∈Zi\in\mathbb{Z}. The difficulty here is the lack of compactness, which we overcome by a discrete version of Lions’s concentration compactness principle from [9].

Superposition of Gibbs measures. Our second result concerns the possibility of finding new Gibbs measures in G(d,Q)\mathcal{G}(d,Q) as superpositions of two non-degenerate Gibbs measures μ,μ′∈G(d,Q)\mu,\mu' \in\mathcal{G}(d,Q). Here, a Gibbs measure μ∈G(d,Q)\mu\in\mathcal{G}(d,Q) is said to be non-degenerate if the corresponding positive solution u∈ℓp(Z)u \in\ell^p(\mathbb{Z}) of (2) is non-degenerate, meaning that the linearized equation

v=d(Q∗u)d−1Q∗vv = d(Q * u)^{d-1}Q * v

References

  1. [1][AHKM24] A. Abbondandolo, F. Henning, C. Külske, and P. Majer, Infinite-volume states with irreducible localization sets for gradient models on trees, J. Stat. Phys. 191 (2024), no. 63.DOI
  2. [2][BKRS26] N. Butler, K. Krishnan, G. Ray, and Y. Spinka, On the local convergence of integer-valued Lipschitz functions on regular trees, Combin. Probab. Comput. 35 (2026), 655–687.arxiv.org/abs/2410.05542
  3. [3]K. Burns and H. Weiss, A geometric criterion for positive topological entropy, Comm. Math. Phys. 172 (1995), 95–118.DOI
  4. [4]K. Deimling, Nonlinear functional analysis, Springer, Berlin, 1985.DOI
  5. [5]H.-O. Georgii, Gibbs measures and phase transitions, Walter de Gruyter, 1988.
  6. [6]F. Henning and C. Külske, Coexistence of localized Gibbs measures and de-localized gradient Gibbs measures on trees, Ann. Appl. Probab. 31 (2021), 2284–2310.DOI
  7. [7]A. Heller and Y. Spinka, Random homomorphisms and Lipschitz functions on trees, arXiv:2606.29426 [math.PR], 2026.arxiv.org/abs/2606.29426
  8. [8]G. James and Y. Starosvetsky, Breather solutions of the discrete p-Schrödinger equation, Localized excitations in nonlinear complex systems: current state of the art and future perspectives (R. Carretero-González, J. Cuevas-Maraver, D. Frantzeskakis, N. Karachalios, P. Kevrekidis, and F. Palmero-Acebedo, eds.), Springer, 2014, pp. 77–115.
  9. [9]P. L. Lions, The concentration-compactness principle in the calculus of variations. The locally compact case, part 1, Ann. Inst. H. Poincaré Anal. Non Linéaire 1 (1984), 109–145.DOI
  10. [10]P. Lammers and F. Toninelli, Height function localisation on trees, Combin. Probab. Comput. 33 (2024), 50–64.
  11. [11]J. E. Marsden and M. West, Discrete mechanics and variational integrators, Acta Numer. 10 (2001), 357–514.DOI
  12. [12]The on-line encyclopedia of integer sequences (OEIS).
  13. [13]R. S. Palais, The principle of symmetric criticality, Comm. Math. Phys. 69 (1979), 19–30.DOI
  14. [14]U. A. Rozikov, Gibbs measures on Cayley trees, World Scientific Publishing, 2013.DOI
  15. [15]M. Struwe, Variational methods, fourth ed., Springer-Verlag, Berlin, 2008.
  16. [16]A. P. Veselov, Integrable maps, Uspekhi Mat. Nauk 46 (1991), 3–45, transl. in Russian Math. Surveys.DOI
  17. [17]S. Zachary, Countable state space Markov random fields and Markov chains on trees, Ann. Probab. 11 (1983), 894–903.DOI

Paper details

vlnc:W_pppmjwde
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 1

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Mathematics

Claude Opus 5.5 (bare), OpenAI GPT-5.4

View statement
Looking for related papers…
Contents